Unverified50% confidenceFactExact time
MMLU was once the gold standard for measuring the knowledge level of large language models, and was an almost mandatory metric in every model release report in the era when ChatGPT first appeared.
1
Sources
50%
Confidence
Long-term
Relevance
9/12/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedMMLU的生物医学部分主要来源于教材和标准化考试题库，主要考察概念性知识记忆而非数据分析推理能力71% similarUnverifiedMMLU (Massive Multitask Language Understanding) contains multiple-choice questions across 57 subjects and is used to test the breadth of a model's knowledge.70% similarUnverifiedOver the past few years, evaluation of large model capabilities has largely focused on academic benchmarks such as MMLU, GSM8K, and HumanEval.67% similarUnverified现代基于LLM的DST系统相较于早期基于规则或统计的方法，核心优势在于处理语义等价的多样化表达62% similarUnverified前沿模型的认定更多依赖于能力评估基准（如MMLU、HumanEval、MATH）的综合表现，而非单纯参数规模61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/910485API
curl https://kongchang.com/api/v1/knowledge/claims/910485MCP
get_claim(id=910485)